This Phase I SBIR project is aimed at developing a novel type of anticancer drugs, termed SNX9- class compounds. These compounds have a dual activity: (i) inhibition of a transcriptional regulatory pathway that endows growth-arrested tumor cells and cancer-associated fibroblasts with tumor- promoting paracrine activities, and (ii) selective killing of tumor cells preferentially to normal cells. The first effect of SNX9-class compounds is prevention of the induction of transcription in cells that are growth-arrested via cyclin-dependent kinase inhibitor (CKI) proteins (the CKI pathway). Proteins upregulated through the CKI pathway include secreted factors that stimulate tumor growth, survival, invasion or angiogenesis. Such activities have been associated with cancer-associated fibroblasts and senescent tumor cells. The second activity of SNX9 class compounds is tumor-selective cytotoxicity. This activity appears to be due to the inhibition of cyclin-dependent kinase 3 (CDK3), an enzyme recently shown to have oncogenic activity, due to activating phosphorylation of a pro- carcinogenic transcription factor ATF1. It is presently unknown whether the two activities of SNX9- class compounds are mechanistically linked. The first goal of this project is to ascertain the role of CDK3 in the tumor-selective cytotoxicity of CDK3-type compounds, by comparing the effects of SNX9 class compounds and CDK3-targeting short hairpin RNA (shRNA) on the growth of tumor and normal cells, cell cycle arrest and induction of cell death. The second goal is to determine if CDK3 and CKI pathway inhibition are mechanistically linked, by analyzing the effects of CDK3 and ATF1 knockdown on the induction of transcription by CKI proteins, and by determining if SNX9-class compounds have the same effect on CKI-CDK interactions as other CKI pathway inhibitors that do not inhibit CDK3. The third goal is to develop and validate an all-atom 3D model of CDK3 for virtual screening of optimized SNX9 derivatives, using high-resolution structures of a closely related kinase CDK2 as a template for ligand-guided homology modeling of CDK3 and its SNX9 interactions. The best model will be used for virtual screening of commercial libraries to identify additional SNX9-related inhibitors. The future Phase II program will include lead optimization of SNX9-class compounds using the principles and homology models developed in Phase I, followed by in vivo safety and efficacy studies of the optimized compounds. PUBLIC HEALTH RELEVANCE: This proposal is aimed at developing a novel class of anticancer drugs, with a unique combination of two beneficial activities. The first activity is killing cancerous cells while sparing normal cells of the body, and the second activity is blocking the production of molecules that promote cancer growth and metastasis. This novel class of drugs has the potential of making significant improvements in the treatment of different types of cancer.